FWIW, I'm still on the earlier side of my career (though certainly not just starting out) and the direction of my learning has been away from fads and towards o

FWIW, I'm still on the earlier side of my career (though certainly not just starting out) and the direction of my learning has been away from fads and towards older, more established technologies. Erlang, Unix, etc. have all been around for many years, and while they do receive active development are still pretty well established and based off of sound engineering principles. I also try and get beyond the hype. REST/HATEOS is cool? What about the things it replaced? What are the edge cases? What was the original design intent? I also read from/talk with older engineers (many of whom I have the privilege to work with) to understand what things were like during previous fads. Ultimately, the problem is about staying marketable while giving fair due to our life outside of work (I prefer the term "unpaid responsibilities" to "leisure time"). I don't claim to have solved this problem, and I still have concerns about what the future holds, but my instinct is that staying on top of fads is a trap that I want to avoid.